[Intel-gfx] [PATCH igt 2/2] igt/perf_pmu: Explicitly reset min_freq before max_freq

Chris Wilson chris at chris-wilson.co.uk
Wed Nov 22 21:16:38 UTC 2017


Quoting Tvrtko Ursulin (2017-11-22 21:07:47)
> 
> On 22/11/2017 19:00, Chris Wilson wrote:
> > The kernel doesn't like it when you set max_freq < min_freq and
> > complains bitterly. So before setting max_freq = min_freq, first set
> > min_freq to min_freq!
> > 
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> > ---
> >   tests/perf_pmu.c | 4 ++++
> >   1 file changed, 4 insertions(+)
> > 
> > diff --git a/tests/perf_pmu.c b/tests/perf_pmu.c
> > index bcb90c84..b8422211 100644
> > --- a/tests/perf_pmu.c
> > +++ b/tests/perf_pmu.c
> > @@ -946,6 +946,8 @@ test_frequency(int gem_fd)
> >       min_freq = igt_sysfs_get_u32(sysfs, "gt_RPn_freq_mhz");
> >       max_freq = igt_sysfs_get_u32(sysfs, "gt_RP0_freq_mhz");
> >       boost_freq = igt_sysfs_get_u32(sysfs, "gt_boost_freq_mhz");
> > +     igt_info("Frequency: min=%u, max=%u, boost=%u MHz\n",
> > +              min_freq, max_freq, boost_freq);
> >       igt_require(min_freq > 0 && max_freq > 0 && boost_freq > 0);
> >       igt_require(max_freq > min_freq);
> >       igt_require(boost_freq > min_freq);
> > @@ -956,6 +958,8 @@ test_frequency(int gem_fd)
> >       /*
> >        * Set GPU to min frequency and read PMU counters.
> >        */
> > +     igt_require(igt_sysfs_set_u32(sysfs, "gt_min_freq_mhz", min_freq));
> > +     igt_require(igt_sysfs_get_u32(sysfs, "gt_min_freq_mhz") == min_freq);
> >       igt_require(igt_sysfs_set_u32(sysfs, "gt_max_freq_mhz", min_freq));
> >       igt_require(igt_sysfs_get_u32(sysfs, "gt_max_freq_mhz") == min_freq);
> >       igt_require(igt_sysfs_set_u32(sysfs, "gt_boost_freq_mhz", min_freq));
> > 
> 
> I don't see any harm, but also don't see how min freq is not min freq at 
> this point. Dealing with random state on the system or some lack of 
> cleanup from the test itself?

In this case, dealing with itself failing at in opportune moments, so
random system state.
-Chris


More information about the Intel-gfx mailing list